What is the burden of proof? Has this principle ever been challenged?

Question

I have been surprised to find that some people doubt this principle. Onus probandi incumbit ei qui dicit, non ei qui negat - the burden of proof lies with the speaker, not with the one who negates. I cite Russell's Teapot and Hitchen's Razor. Occam's Razor provides context to the argument. I have seen no convincing argument against this principle.

It's mostly just a principle of debate, not so much of reality. And what is the burden of proof for the belief in the burden of proof? — Kristian Berry, Nov 12 '23 at 14:27
You seem to believe that there is something called "the burden of proof," no? But does the principle of this burden, apply to itself? If not, why not? (I realize we approach the regress of knowledge, here, of course...) — Kristian Berry, Nov 12 '23 at 15:45
I have no such belief. What makes you think that? A principle is a tool. If I questioned a spanner, would that make sense? Btw, why are you posting alleged answers in comments? Any chance of an actual answer? — Meanach, Nov 12 '23 at 17:42
I would have imagined that the purpose of arguments, when they are used, is to generate belief in their conclusions. So if there is an argument that the phrase "burden of proof" has substantive and useful meaning, I would think the outcome of accepting the argument would be accepting, and hence believing in, the conclusion (that there is such a thing as a burden of proof). As to why I sometimes post what look like answers in the comments, sometimes I am discouraged because I have posted lengthy citation-based answers to questions and seen them get little traction. — Kristian Berry, Nov 12 '23 at 18:50
Of course, at other times my answers get a decent number of upvotes and are even accepted by the initial inquirer, so perhaps I am being quasi-cynical when I refrain from trying for acceptable answers rather than illustrative comments (think of the comments as similar to objections that might be raised during peer review of a dissertation, anyway...). — Kristian Berry, Nov 12 '23 at 18:52
Like most things, the expression "burden of proof" got abused the minute it was coined as an expression, so it shouldn't be too surprising that "some people doubt this principle". — Stef, Nov 12 '23 at 23:27
“Ockham’s Razor” isn't directly related to *burden of proof. It's a different consideration. Whether you're the side with the putative burden of proof or the other side, all William of Ockham is saying is that simple explanations that are sufficient to explain are preferable to complicated explanations that also suffice to explain. Einstein had a Lemma to Ockham: "Things should be as simple as possible, but no simpler."* — robert bristow-johnson, Nov 13 '23 at 02:46
Indeed, in the middle ages it was suite customary to torture people, in expectations that, of not guilty, they would not confess of a crime. In other words, they had to prove their innocence. Of course, many confessed just to stop being tortured. — Roger V., Nov 13 '23 at 11:07
Indeed. James I abolished torture because it produced unreliable evidence. Btw, I did not claim that Occam's Razor spoke directly to the burden of proof. It is background to the argument. — Meanach, Nov 13 '23 at 11:13
I guess if someone didn't like the Burden of Proof method, they would have to come up with something better. Otherwise, they would simply unravel the nature of argumentation, and, lacking anything to replace it with, we would put it back. People aren't good at just throwing their hands in the air and giving up. — Scott Rowe, Nov 13 '23 at 11:23
A burden of proof is an ought. There is no fact that specially delimits one ought from another. Oughts do not come from is statements, as Hume showed. As such, it is not just that it can be challenged. A burden of proof, by any objective measure, cannot exist. It is nothing more than a subjective claim of what one “feels” is right in a debate context. — , Nov 12 '23 at 15:38
If someone claims X, they should* have a rational reason for doing so. If you claim not-X, you should* have a rational reason for doing so. The burden of proof is on both parties. Just because someone previously claimed X before you claimed not-X does not free you from your obligation* to justify your belief in not-X. Whatever you believe, you should* have a rational justification for, except for those things so intrinsic to your reasoning that they serve as axioms. However, there is often a good, default reason for believing not-X: most conceivable claims that a specific thing exists are not — causative, Nov 12 '23 at 16:15

score 10 · Accepted Answer · edited Nov 13 '23 at 23:20

The burden of proof falls on the one who makes the claim because, usually, they don't make this claim in a vacuum, without any goal in mind: they want other people to accept their claim and adjust their behavior accordingly.

"I believe people who wear corduroy pants offend God and I want you to stop wearing those."
"I think project X is a waste of resources and you should stop working on it."
etc.

Imagining a person making a claim without any specific goal, or without wanting other people to believe the claim is kind of difficult. It does not make sense; why make the claim in the first place? But if such a case were to exist, the burden of proof would not be an issue. "You believe the Earth is hollow, but don't care if I think the same and don't want me to do anything about it? Cool. You do you."

There is also the special case of claims with very small stakes, like "Hello, my name is Mike". The claimant clearly expects their vis-a-vis to adopt the behavior of calling him "Mike", but people rarely ask for ID to prove such a claim, because the stakes are very low.

If one wants people to change their behavior based on a claim, it's only fair that this person does the legwork to convince whoever they are addressing the claim to. In other words, the one who requires the change in behavior has to have people consent to this change. Otherwise the only option left to have them comply is force, i.e. potentially tyranny.

A decent argument could be made for this principle based on various moral theories, like the Golden Rule, Kant's Categorical Imperative, and Social Contract theory.

Interesting example. The evidence for "Mike's" name was the event of him saying it, because a person knows their own name. Perhaps Mike was lying, but a polite person would not assume, accuse, or even ask — unless they are given evidence that taints the person's character or reputation. Basically, lying complicates the burden of proof. — Matthias, Dec 12 '23 at 05:01
@Matthias true. Claims about personal thought or state of mind can also be seen as both claim and evidence about the claim, like in "I feel sad" or "I think corduroy pants offend God". — armand, Dec 12 '23 at 06:46

score 4 · Answer 2 · answered Nov 12 '23 at 16:27

The "burden of proof" is the principle that if you have a good reason to believe what you believe, and you want me to believe what you believe, you ought to say your good reason to believe what you believe, and not rest your argument on the demand that your interlocutor explain why you shouldn't believe what you believe.

Saying, "You have the burden of proof, not me!" in place of an explanation of your good reason to believe what you believe... is resting your argument on the demand that your interlocutor explain why you shouldn't believe what you believe.

Saying, "You have the burden of proof, not me!" is shifting the burden of proof. There *is* a legitimate discussion about when such shifting is legit and when it is not legit. That's the whole argument about *who* has the burden of proof regarding any particular subject. — robert bristow-johnson, Nov 13 '23 at 03:00

score 3 · Answer 3 · answered Nov 12 '23 at 18:49

The burden of proof is merely the result of trying to figure out what beliefs we should start with and what should be proven from there.

It stops you from believing anything and everything without question, and being consistent in your reasons for believing things.

Consider the legal system: if someone is suspected of murder, we don't just throw them in prison unless they can prove they're innocent. The burden is on proving guilt.

There are justice-related reasons for this. But also, functionally speaking, it may be somewhere between difficult and impossible to prove innocence. How would you even be able to prove that at no point did you kill someone? You may need to account for every second of your time over days/months/decades. Whereas proving guilt merely needs to establish what happened at one particular moment in time.

And if 2 people are independently suspected of murdering a particular individual, do we just throw both in jail? They couldn't both have independently murdered the same person (putting aside from fringe cases where someone's heart stops, is restarted, and then stops again, although it's arguable whether the first case would then be murder). Concluding that they did so would be contradictory, and this could trivially be extended to 3 people, or 4, 5, 100, a million or 8 billion.

Similarly, if someone proposes the existence of some being, we shouldn't just accept that it exists without question. Proving non-existence is somewhere between difficult and impossible (combing every inch of the planet to prove that Bigfoot doesn't exist), and the existence of different beings may contradict one another. The burden should be on proving existence.

The idea of the burden of proof is almost universally accepted in some capacity (whether it's accepted by that name or not). It's pretty fundamental to how we learn things about reality. If you're in the forest, it wouldn't be smart to just eat random plants assuming they're safe to eat, but rather you should stick to ones you already know to be safe (if any) and look for proof that others are safe before eating them.

When people try to argue against it, they don't tend to argue against the principle as a whole (or they argue against a strawman of it), but rather they argue that a particular claim or set of claims shouldn't be subject to the burden of proof or that the burden of proof should be on the opposite claim. This is common in theistic discussion, where atheists ask for proof of God's existence, but are often instead asked to prove God's non-existence, or where theists say that God's existence doesn't need to be proven, that it's something that everyone already believes, or that it's a presupposition.

Maybe there's also some corner of philosophers who reject the idea of the burden of proof altogether (maybe those who don't believe in objective reality). Although I might challenge anyone who claims as much (unless they have almost zero autonomy), because I suspect you'd have a hard time surviving for more than a few days without that.

Note: "proof" in "burden of proof" doesn't mean a conclusive, 100% reliable and exhaustive mathematical or logical proof (which isn't possible for claims of reality in any case). It merely means sufficient evidence or other justification to warrant accepting a claim.

I keep disbelieving in objective reality, and existence keeps proving me wrong. But, it gives me something to do. — Scott Rowe, Nov 13 '23 at 11:52

score 2 · Answer 4 · answered Nov 12 '23 at 14:41

2

The burden of proof is the fundamental tenet of skepticism. It insists that there is a burden to prove, and that is not accepted by many people.

I have seen no convincing argument against this principle.

Yes you have, and it's called faith. Faith is a guiding principle for far more people currently and throughout history than doubt. And in fact, it can be extended beyond theological arguments. Philosophy explores this under the topic 'fideism'. From WP:

Fideism (/ˈfiːdeɪ.ɪzəm, ˈfaɪdiː-/) is a term used to name a standpoint or an epistemological theory which maintains that faith is independent of reason, or that reason and faith are hostile to each other and faith is superior at arriving at particular truths (see natural theology). The word fideism comes from fides, the Latin word for faith, and literally means "faith-ism".1 Philosophers have identified a number of different forms of fideism.2 Strict fideists hold that reason has no place in discovering theological truths, while moderate fideists hold that though some truth can be known by reason, faith stands above reason.

In my own upbringing, the chief argument of faith was presented as Jesus's exhortation to be as children and submit to the love of God.

And he said: “Truly I tell you, unless you change and become like little children, you will never enter the kingdom of heaven. Therefore, whoever takes the lowly position of this child is the greatest in the kingdom of heaven. And whoever welcomes one such child in my name welcomes me.” - Mathew 18:3-5

It is a call to humility to submit to the will of God, and is American as apple pie in the sense that fundamentalist Christian refugees populated the East Coast starting in the 17th century and continue to resist the call to reason. Prosperity theology in the US thrives on rejecting the skeptical burden of proof for better or worse.

That argument may not convince you, but the argument is alive and well and is accepted by billions of people.

answered Nov 12 '23 at 14:41

J D

26,214
3
23
98

2

I'd add that the freedom to doubt God is a relatively newly sanctioned social rite in Western Europe, and the reason why the Freemasons were a secret society for a long time was because Occam's Razor was often seen as treason against the King and the kingdom. – J D Nov 12 '23 at 14:47
This is not an argument. But have it your way – Meanach Nov 12 '23 at 14:59
@Meanach On the narrow reading of the argument, true enough. But an argument is a rhetorical speech act, not a logical structure expressed in a concrete syntax. The latter view is the narrow view, and a wise man understands and uses both definitions of argument. – J D Nov 12 '23 at 15:05
Oh, and if you have any doubt, just review Philosophical Investigations and the notion of the language-game. It prefigures the modern psychologism's interpretation of logic and reason and the normativity that inheres to all language. It will make it easier to deal with understanding intersubjectivity, theory-laddenness, and the basis for accepting non-classical logic when you get to those topics. – J D Nov 12 '23 at 15:10
I would be interested in an actual answer that uses Wittgenstein to justify theism. – Meanach Nov 12 '23 at 15:27
@Meanach No, of course "You have to take it on faith, faith is a good thing" is not a sound argument against the principle. But it is — definitely — a challenge to the principle. – MichaelK Nov 12 '23 at 15:57
How so? Perhaps you might post an answer? – Meanach Nov 12 '23 at 17:38
3

Faith is not an argument against the burden of proof. At most it's merely stating that someone doesn't apply the burden of proof to some claims. Faith tends to just be asserted to be part of one's epistemology, rather than people trying to justify why faith should be used. – NotThatGuy Nov 12 '23 at 17:40
1

@NotThatGuy Yeah, you'd be right except for the fact for 1,000 years, there's an entire body of argumentation of why faith is an argument against the burden of proof. You reject the conclusion of fideists, but it's not an open debate on whether one can make arguments for it. I'm an atheist, so I give little credence to faith. But that doesn't entitle me to conclude that the logical arguments supporting the adoption of faith are meaningless or non-existent. They just don't persuade me, and I'd be an ass to dismiss all of theology as irrational. – J D Nov 12 '23 at 18:00
@JD Faith is not an argument any more than "this apple is red" is an argument against "this apple is blue". It's merely an opposing position. You may be able to make an argument for why the apple is red (why you should use faith), but simply asserting that it's red is not an argument. And, like I said, most people just assert that faith should be used without an argument for why. Well, okay, many say a book says so, which I'll concede could technically be considered an argument, it would just be a very poor one. – NotThatGuy Nov 12 '23 at 18:10
@NotThatGuy I never said faith was an argument. I said theologians put forth arguments for faith. When you get that straight, you'll understand that my post does not endorse faith, but merely concedes that there are arguments for and against faith, and the rejection of faith without consideration of arguments for it is a bias, not a well-thought out position. – J D Nov 12 '23 at 20:20
@JD "I never said faith was an argument" - you said "faith is an argument"... and said the same in your answer, in different words ("[a convincing argument is] called faith"). But I didn't say you consider faith to be a good or sound argument. If you accept it isn't an argument, then you might want to edit your answer appropriately. This may sound like nitpicking, but most people who appeal to faith seem to think simply saying "I have faith" is a sufficient argument for believing something without sufficient evidence, but in reality it's merely asserting that they hold such a belief. – NotThatGuy Nov 12 '23 at 20:31
1

@NotThatGuy To be fair, and I read from stem to stern, I concede did use the phrasing. "faith is an argument" which was intended as an ellision of "(Have) Faith (because X) is an argument." I played fast and sloppy with both senses of faith (as an epistemological disposition and as a conclusion by way of fideism) in my construction, and I apologize for the confusion. – J D Nov 12 '23 at 21:45
I'm willing to accept that reason cannot reach to some claims. However, I don't want little children designing airplane engines, power grids and railway control systems, performing surgery... – Scott Rowe Nov 13 '23 at 12:51

score 2 · Answer 5 · answered Nov 12 '23 at 17:40

2

Let us take an interesting example , suppose your neighbour claims he is your Dad ,then where does the burden of proof lies? On your neighbour ,or your mother ,or on you ,or your present Dad ? The situation is very tricky. Once a doubt in relationship has been created , you would like verify it yourself by asking your mother. It is also possible that your neighbour might like you to verify the claim by doing DNA testing. Burden of proof is a subjective prerogative. It depends on the situation.

answered Nov 12 '23 at 17:40

Dheeraj Verma

1,851
2
11
16

If things are that murky, maybe you should not inquire further? – Scott Rowe Nov 13 '23 at 11:54
@ScottRowe yes. But it will always fluster you. Wouldn’t it ? – Dheeraj Verma Nov 13 '23 at 12:10

Jo Wehler · Answer 6 · 2023-11-13T12:30:36.507

The rule of onus probandi states that the burden of proof lies with the speaker, not with the one who negates, see the OP's question.

The three other principles mentioned in the OP’s question are characterized by wikipedia as follows:

"Russell's teapot is an analogy, formulated by the philosopher Bertrand Russell (1872–1970), to illustrate that the philosophic burden of proof lies upon a person making empirically unfalsifiable claims, rather than shifting the burden of disproof to others”, see https://en.wikipedia.org/wiki/Russell%27s_teapot

“Hitchens's razor is an epistemological razor that serves as a general rule for rejecting certain knowledge claims. It states "what can be asserted without evidence can also be dismissed without evidence."”, see https://en.wikipedia.org/wiki/Hitchens%27s_razor

“Occam's razor [...] is the problem-solving principle that recommends searching for explanations constructed with the smallest possible set of elements”, see https://en.wikipedia.org/wiki/Occam%27s_razor

All of the three principles are not laws of nature. They cannot be detected by research. Instead, they are principles of debate as @KristianBerry points out. Therefore they have to be negotiated.
The principles of Russell’s teapot and Hitchens’ razor do not imply that it is always the person, who makes a claim of existence (proponent), who has the burden of proof (rule of onus probandi), but not his opponent.
The rule of onus probandi breaks the symmetry between the proponent and the opponent, because it shifts the burden of proof to the proponent.

A fair debate should provide some argument, why to follow this rule, and should secure the agreement of the interlocutors to accept the rule of onus probandi.

An argument for accepting the rule of onus probandi in a debate between theism and atheism could be: The proponent can prove his claim of existence just by presenting the one entity. While the opponent in general cannot show the non-existence of the entity on all locations.

The asymmetry is between showing the existence on one location and showing the absence on all locations. The only chance of the opponent is to detect a logical contradiction in the argument of the proponent.

Hence, the evidence of logical inconsistency in the claim of the theist is one of the strongest arguments in the hand of the atheist. See Leibniz’ attempt to argue against the criticism uttered against the theistic position.
If there is agreement that the premisses of Occam razor are fullfilled, then it seems reasonable to prefer the more simple theory: Why make it complicated when you can make it simple?

score 1 · Answer 7 · answered Nov 13 '23 at 11:19

1

The burden of proof is on the proposer of the argument. There is no false equivalence. Principles are tools. It is pointless to discuss their "reality".

answered Nov 13 '23 at 11:19

Meanach

2,337
7
36

score 0 · Answer 8 · answered Nov 13 '23 at 05:01

I think burden of proof should be understood in a context of a claimant versus a skeptic. The claimant, taking initiative, says X is necessarily true. The skeptic says that X may possibly be true, but not-X may also possibly be true. The burden of proof is on the claimant to convert the skeptic from a position of uncertainty to one of certainty.

For example, take presumption of innocence in the U.S. legal system. The burden of proof is on the prosecution to convert a skeptical jury from uncertain of guilt to certain of guilt. The defense is not required to convince the jury of innocence. Nor, despite the term "presumption of innocence", is the jury expected to initially believe the accused is certainly innocent; the jury starts out uncertain.

I think "burden of proof" is more controversial and less applicable when it involves trying to change someone's mind from a "status quo" position of certainty to an opposite one. I.e. the claimant says X, and the skeptic currently believes not-X. You could argue the claimant has the burden of proof because she is trying to change the status quo, but you could argue that the skeptic is also claiming a position, not-X, which also requires proof. In this case there could be a lot of arguing over which stance is actually the "status quo" and who has the burden of proof.

Ok. But arguing over which position is the status quo seems to indicate that there's not actually a status quo, just ignorance all around. Everyone has to support their position in that case. Don't stop until it is blindingly obvious to everyone. — Scott Rowe, Nov 13 '23 at 11:37
@ScottRowe agreed, which is why I wrote that "burden of proof" is less applicable to that kind of situation. — usul, Nov 14 '23 at 02:24

score -1 · Answer 9 · edited Nov 13 '23 at 21:23

It's useful to ask the question outside the realm of theology. We accept many things on the basis of trust: for example, the fact that a hydrogen atom has one proton and one electron, or the fact that Britain is an island, or the fact that vaccines confer immunity. We believe these things primarily because there is no reason to doubt the veracity of the claim, and we have confidence that the claim could be proved, or has been proved, even though we haven't seen the proof and might not understand it.

But if there is reason to doubt the claim, for example, because we don't trust the person making the claim, or because other people claim the opposite, then it's reasonable to argue that we shouldn't believe the claim unless the person making the claim can prove it to our satisfaction: at that point, the burden of proof falls on them.

There is copious proof that vaccines confer immunity. Your other examples are also easily verified. Flat earthers can take a flight and observe the curvature of the Earth. Your answer is factually incorrect. I agree that it does not only apply to theism. — Meanach, Nov 13 '23 at 10:45

What is the burden of proof? Has this principle ever been challenged?

9 Answers9

Linked